Disputed Points in Teleostean Emhryohgy. 217 



relation exists in Petromyzon ; but he merely sajs that the 

 perivitelline space is partcf the segmeutation-cavity, and that 

 it subsequently becomes shut off by the downgrowth of the 

 mesoLlast, and forms the subintestinal vein. He makes no 

 comparison between the perivitelline blood- sinus and the 

 vitelline blood-vessels of other forms. 



If M^ntosh and Prince had really understood the later 

 history of the segmentation-cavity they would never have 

 argued that it was the gastriila-cavity. The gastrula-cavity 

 must by its definition become the lumen of the intestine, and 

 the segmentation- or " germinal " cavity never has any con- 

 nexion with the lumen of the intestine. I have shown that 

 the real representative of the gastrula-cavity in Teleosteans 

 is Kupfi'er's vesicle. 



liyder in 1884 (18) was of opinion that the segmentation- 

 cavity in the later stages was " synonymous " with the body- 

 cavity; whether he still holds this opinion I do not know, 

 but it is evident from the above that it is entirely erroneous ; 

 the segmentation-cavity has as little to do with the body- 

 cavity as it has with the gastrula-cavity. The apparent 

 continuity in development of the segmentation-cavity with 

 the perivitelline blood- sinus is due entirely to the retarded 

 development of the mesoblast in pelagic ova and certain 

 others ; while the obliteration of the segmentation-cavity by 

 the mesoblast, which takes place in Amphibian and many 

 Teleostean ova, is represented in pelagic Teleostean ova by 

 the formation of mesoblastic cells from the periblast. The 

 cavity ceases to be a segmentation-cavity and becomes a peri- 

 vitelline blood-sinus as soon as any definite mesoblastic cells 

 are produced on its inner wall. 



The account given by M'Intosh and Prince (9) of the 

 development of the heart is, as a whole, to me quite incom- 

 prehensible, while many of the separate statements in that 

 account are, I venture to say, erroneous. One of these state- 

 ments is that '' the heart usually pushes down before it a 

 delicate stratum of hypoblastic cells ; but this limiting ventral 

 layer apparently becomes obliterated anteriorly, and the peri- 

 cardial chamber is open to the subembryonic space, which is 

 undoubtedly the persisting germinal cavity." This is the 

 most extraordinary confusion. What has the hypoblast to 

 do with the formation of the heart ? Is it conceivable that 

 the germinal cavity, which, according to these authors, is the 

 gastrula or intestinal cavity, can be open to tlie pericardial 

 cavity ? Of course it is known that in certain forms the 

 original gastrula-cavity segments off portions vviiich form the 

 body-cavity — Antphiuxus^ tor instance. But in Teleosteans 



Ann. & Mag. N. Hist. Ser. 6. Vol. vii. 15 



